help button home button ClinMed NetPrints
HOME HELP FEEDBACK BROWSE ARTICLES BROWSE BY AUTHOR
Warning: This article has not yet been accepted for publication by a peer reviewed journal. It is presented here mainly for the benefit of fellow researchers. Casual readers should not act on its findings, and journalists should be wary of reporting them.

This Article
Right arrow Abstract Freely available
Services
Right arrow Similar articles in this netprints
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Back, I. N
Right arrow Articles by Finlay, I. G
Right arrow Search for Related Content
PubMed
Right arrow Articles by Back, I. N
Right arrow Articles by Finlay, I. G
Related Collections
Right arrow Anaesthesia:
Pain

Right arrow Other Anaesthesia
Right arrow Oncology:
Cancer:other

Right arrow Other Oncology
Right arrow CLINICAL:
Palliative Medicine

clinmed/2000070007v1 (August 15, 2000)
Contact author(s) for copyright information

Caudal epidural analgesia in palliative care: a review of case notes.

Dr I. N. Back, MA MRCGP DA
Consultant in Palliative Medicine

Professor I. G. Finlay, FRCGP
Professor of Palliative Medicine

Correspondence to: Dr I. N. Back
Holme Tower Marie Curie Centre, Bridgeman Road, Penarth, South Glamorgan CF64 3YR
Tel: 029 2042 6000; Fax: 029 2042 6036; E-mail: ian.back{at}dial.pipex.com

Abstract

During three years, 41 patients were treated with caudal epidural injections in a palliative care setting. Thirty nine patients received 65 caudals containing corticosteroid for malignant pain, and their case-notes are reviewed. Response to the injection, as judged from the case-notes was 'good' in 36 cases (56%), 'some' in 16 (25%), and 'none' in 12 (19%); response was not recorded in one patient. The response in patients with pain limited to areas supplied by sacral nerves showed a higher percentage of 'good' responses, 68% versus 41% (p<0.05) than for those with pain also in areas supplied by lumbar nerves. Two patients had caudal injections of local anaesthetic alone to provide anaesthesia for manual evacuation of impacted faeces, with excellent results. The caudal approach to the epidural space is discussed. It is suggested that caudal corticosteroid injections are a useful treatment for mixed malignant pelvic pain, and that caudal anaesthesia can provide excellent conditions for manual evacuation of impacted faeces.

Introduction

Lumbar epidural injections of corticosteroid and local anaesthetic are effective in treating radicular pain from non-malignant diseases of the lumbar spine (Spaccarelli 1996), and have been used in palliative care for neuropathic pain of malignant origin (Wood et al. 1992). Most units rely on anaesthetist colleagues to perform epidural injections, as it is a technique not easily learnt outside regular anaesthetic practice.

The caudal approach to the epidural space is commonly used by anaesthetists for peroperative analgesia, especially in children (Dalens and Hasnaoui 1989; Splinter et al. 1995). In a caudal injection, a standard hypodermic needle is inserted through the sacrococcygeal ligament which lies across the sacral hiatus, into the caudal canal which is continuous with the lumbar epidural space. It is technically easier to perform than a lumbar epidural injection, and there is a decreased risk of dural puncture because the dural sac in adults usually ends at the second sacral vertebra, three spinal segments above the sacral hiatus (Bromage 1978). If sufficient volume is injected then drugs will reach the lumbar region, but the approach can also be used specifically to apply drugs locally to the sacral nerve roots of the cauda equina.

Some authors advocate the use of fluoroscopy to confirm needle placement, but radiographic equipment is rarely available in hospices, outpatient clinics or patients' own homes.

Both authors have been using caudal epidural injections without radiological imaging, as well as lumbar epidurals for several years for patients with neuropathic pain in the regions supplied by sacral and lower lumbar nerve roots. We have undertaken a review of the case-notes to look at the efficacy and complications.

Methods

Palliative care patients are seen in three centres, two palliative care units and the regional radiotherapy centre; caudal injections are performed in in-patient, day care and out-patient settings. Patients who received caudal injections were identified by searching the computer databases in each centre. In two of the centres, caudal injections are recorded as coded events; in the other, clinic annotations are held on computer, and the annotation text fields were searched for the word 'caudal'.

Sixty eight references to caudal injection in 42 patients were identified between September 1993 (the earliest computer records) and September 1996. In one patient the injection was recorded as unsuccessful as the sacral hiatus was not identified. The case-notes of the remaining 41 patients were independently reviewed by a retired consultant in Public Health Medicine who holds an honorary post at one centre. Details of the patients, the indication for injection, site and nature of the pain, previous pain treatments, and drugs used for the injection were noted. Efficacy was judged when possible from the case-notes as reported by the patient as none, some, or good. Any complications were also noted.

Two of the patients had a caudal injection to provide anaesthesia for the manual evacuation of impacted faeces, and data on these patients are commented on separately. The remaining 39 patients received 65 injections for pain control.

Details of the 39 patients are given in table 1. The age range was 37-86 years (median 61 years). All patients had malignancy as their primary diagnosis. Twelve patients had more than one injection, with a maximum of five injections in one patient. (Table 2)

Results

Data for the two patients who had caudal anaesthesia for manual evacuation of impacted faeces were looked at separately. Local anaesthetic was used on its own in both cases, 20ml lignocaine 1%, or 10ml bupivicaine 0.5%. In both cases, severe faecal impaction required manual evacuation, which had previously failed under sedative analgesia. Excellent anaesthesia was achieved in both cases, with good anal sphincter relaxation, making faecal removal easier.

The indication for the caudal injections in the remaining 39 patients was pain control. Seventy-four percent of the patients had pelvic tumour, most commonly local recurrence of rectal carcinoma; four patients (10%) had perineal or perianal disease, and 13 (33%) had bone metastases in the lumbo-sacral spine. The documented sites of pain are shown in table 3. Forty-four percent of patients had sites of pain limited to areas supplied by the sacral nerve roots. The outcome of injections given to these patients was compared to those who also had evidence of pain in areas supplied by the lumbar nerve roots, in view of the unpredictable spread of caudal injections to the lumbar epidural space.

The majority, if not all the patients, probably had some element of neuropathic pain, but it was not possible to determine the exact nature of many patient's pain from the notes. Pain was described as burning in five, shooting in two, and lancinating in one patient. Others had recorded evidence of altered sensation in the perineum , or pain that radiated down the leg.

Previous therapies were determined as far as possible from the notes, but are unlikely to be complete. The majority of patients (85%) were receiving strong opioid analgesics, 16 (41%) were taking NSAID's, 19 (49%) were taking a tricyclic antidepressant, and five (13%) were taking anticonvulsant drugs. Twenty four patients (62%) had received radiotherapy to the pelvis or sacrum. The number of other therapies previously tried for pain, including an epidural (one patient), ketamine (1) and prednisolone enemas (1) are some indication that these were difficult pains to control.

All patients received a mixture of bupivacaine, in doses ranging from 5ml of 0.25% to 20ml of 0.5%, with Depomedrone (methylprednisolone acetate) 40mg to 80mg. Thirty five injections (54%) used 10ml of 0.5% bupivacaine, and 59 injections (91%) used Depomedrone 80mg.

Response to the injection was judged as 'good' in 36 cases (56%), 'some' in 16 (25%), and 'none' in 12 (19%) (see table 4). Injections in patients with pain limited to areas supplied by sacral nerves showed a higher percentage of 'good' responses, 68% versus 41% (p<0.05, chi-squared=4.56, df=1).

Length of effect was not recorded in the case of 16 (25%) successful injections; the range of recorded response was 1 day to 8 weeks, with a median of 7 days. (See table 5)

The notes were carefully reviewed for any record of complications. The only recorded possible complication was one patient who reported a 'flu-like' reaction following her fourth injection. She reported the same reaction to a later increase in her dose of octreotide, and subsequently responded well to antidepressant medication.

Sixteen of the patients are still alive at the time of writing, and 25 have died. The interval from the caudal injection to death ranged from 2-453 days (median 124 days), with five deaths occurring less than 2 weeks after the injection. There was no indication from the case-notes that the injection influenced any patients' survival. All patients had advanced metastatic cancer.

In one patient in whom depomedrone was used, increased appetite and sense of well-being followed the first injection. This response paralleled the good response to pain, and was reproducible on two further occasions, each after about three to four weeks.

Discussion

This review relied on a retrospective analysis of case-notes. The results are therefore open to interpretation regarding their validity. Nevertheless, in view of the lack of any other published data on caudal injections in palliative care, we felt it was important to collate and report our experience.

Our review suggests that caudal injections of corticosteroid are effective adjuncts in the management of malignant pain. The success rate of 56% (good response) or 81% (good or partial response) compares well with the overall success rate of 65% reported in Spaccarelli's recent review of epidural corticosteroid injections (Spaccarelli 1996). The majority of our patients probably had mixed neuropathic/nociceptive pain affecting sacral nerve roots, with or without lumbar nerve root involvement. Forty-four percent had carcinoma of the rectum, in which local recurrence commonly results in a mixture of visceral pain, sacral bone pain, and neuropathic pain from sacral nerve root infiltration.

It is not surprising that the efficacy of the injections was significantly higher in those patients with pain only in areas supplied by sacral nerve roots, as the average volume of injection of 10ml would be unlikely to give consistent spread to the lumbar region.

Published studies of caudal injections of corticosteroids have mostly been in non-malignant pain in the lumbar spine associated with radicular pain, especially sciatica (Spaccarelli 1996). In these patients the caudal approach may be less useful because spread of the injected fluid to the lumbar region is unpredictable, and the variable anatomy of the sacrum makes needle placement less predictable. However, the caudal route is advantageous in patients who may have metastatic disease in the lumbar vertebrae, or find it too painful to adopt the position needed for a lumbar epidural; there is also a decreased risk of dural puncture.

Needle misplacement may occur in as many as 52% of cases when inexperienced doctors perform caudal injections (Renfrew et al. 1991). In one study 25% of caudals were shown to be incorrectly placed using fluoroscopic demonstration techniques (White et al. 1980); however, in the same study 30% of lumbar epidurals were also misplaced. White concludes that needle misplacement is likely to occur in less than 5% of cases with a cooperative patient when the sacral hiatus is easily identified (White et al. 1980). There are no reports to suggest that injection outside the epidural space has lead to any serious complications.

No serious complications were noted following any of the injections in this study. Only a few case reports have described major complications from caudal or lumbar epidural corticosteroid injections, and the actual incidence is unknown (Spaccarelli 1996). Respiratory arrest due to accidental subarachnoid injection (dural puncture) is a well recognised complication following lumbar epidurals, but has not been reported following caudal injection in adults. Seizures or cardiac arrest could occur if a large dose of local anaesthetic were inadvertently injected intravenously, but the true degree of risk is difficult to ascertain.

Semple and Bisset (Semple and Bisset 1985) report the loss of consciousness following a caudal injection and suggest that this may be due to a transient rise in intracranial pressure caused by a rapid rate of caudal injection; this may be more likely to occur in patients with raised intracranial pressure due to intracerebral pathology.

Dalens and Hasnaoui (Dalens and Hasnaoui 1989) reported a 10% incidence of intravenous placement of the needle in a study of 750 caudal injections in children. Most of these were detected by aspiration of blood prior to injection. Even after resiting, 3 (0.4%) inadvertent intravenous injections occurred; these were detected following a small test dose containing adrenaline by monitoring cardiac dysrhythmias.

In fluoroscopic studies, Renfrew (Renfrew et al. 1991) reported a 9% incidence of injections into an epidural vein, and White (White et al. 1980) reported a 6% incidence. Considering the large number of caudal injections performed in anaesthetics and the paucity of reports of cardiac arrest or seizure, it seems unlikely that many of these inadvertent intravenous injections, if they occur, lead to serious complications. The use of lignocaine instead of bupivicaine could reduce the risk of cardiac dysrhythmias as it is less cardiotoxic (Anonymous 1986).

The length of response in our series is perhaps disappointing, with 18 of the effective injections giving relief for less than 8 days. However it should be noted that 16 of the effective injections had no record of the length of action. The nature of the retrospective case note review biases against finding records of the effect wearing off, the longer it is after the injection. In White's study of 300 patients given epidural corticosteroid injections for low back pain (White et al. 1980), 82% of patients reported pain relief after 24 hours, 50% after two weeks, and 24% after one month.

The two cases of caudal anaesthesia with local anaesthetic alone, used for manual evacuations of impacted faeces are very well remembered by the nursing staff of the unit. In addition to very effective analgesia, reduction of anal tone made the procedure much easier.

We conclude that the caudal epidural injection is a useful technique in palliative care, to administer corticosteroid in cases of malignant pain in the pelvis, and local anaesthetic for painful procedures. Despite the difficulty in ensuring correct needle placement, it is a procedure that can be learnt more easily than lumbar epidural injection. There is a risk of cardiac or respiratory arrest occurring due to unintentional intravenous injection, but this appears to be rare, and must be weighed up as always with the potential benefits and risks of other treatments.

Acknowledgements

The authors gratefully acknowledge the assistance of Dr Neville Hughes MBE, for his help with the review of the case-notes, and Dr Robert Newcombe for his statistical advice.

References

 

TABLES

Table 1 Patient details

 

 

No. patients

Age (years)

<50

7

 

50-59

10

 

60-69

12

 

70-79

7

 

>=80

3

Sex

Male

17

 

Female

22

Diagnosis

Ca. Rectum

17

 

Ca. Colon

5

 

Carcinoid of caecum

1

 

Ca. Anus

2

 

Ca. Breast

3

 

Ca. Vulva

3

 

Ca. Uterus

1

 

Ca. Cervix

1

 

Ca. Ovary

1

 

Ca. Prostate

2

 

Ca. Bladder

1

 

Ca. Skin (SCC ear)

1

 

Lymphoma

1

Total

 

39

Table 2 Number of caudal injections given

No. caudal injections

No. patients

1

27

2

4

3

4

4

2

5

2

Table 3 Sites of pain

Site of pain

No. patients

 

Perineal

16

41%

Rectum

8

21%

Sacral

5

13%

Pelvic

5

13%

Anus

4

10%

Sacro-iliac

2

5%

Suprapubic

1

3%

Legs

18

46%

Lumbar spine

7

18%

Groin

3

8%

Lumbo-sacral

2

5%

Loin

1

3%

Testis

1

3%

Sacral nerve roots only

17

44%

Mixed

22

56%

Total

39

100%

Table 4 Efficacy on pain

 

No. Injections (%)

Effect:

None

Some

 Good

Pain confined to sacral nerve distribution

5

(14%)

7

(19%)

25

(68%)

Pain not confined to sacral nerve distribution

7

(26%)

9

(33%)

11

(41%)

Total

12

(19%)

16

(25%)

36

(56%)

(1 injection - no record of effect)

Table 5 Length of effect of injections

Duration of effect

Effect

 

 

None

Some

Good

Total no. patients

No effect

12

-

-

12

1-3 days

-

6

5

11

4-7 days

-

2

5

7

8-14 days

-

2

8

10

3-4 weeks

-

1

4

5

5-6 weeks

-

-

1

1

7-8 weeks

-

-

2

2

Duration not recorded

-

5

11

16

(1 patient - effect not recorded)

 

 





This Article
Right arrow Abstract Freely available
Services
Right arrow Similar articles in this netprints
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Back, I. N
Right arrow Articles by Finlay, I. G
Right arrow Search for Related Content
PubMed
Right arrow Articles by Back, I. N
Right arrow Articles by Finlay, I. G
Related Collections
Right arrow Anaesthesia:
Pain

Right arrow Other Anaesthesia
Right arrow Oncology:
Cancer:other

Right arrow Other Oncology
Right arrow CLINICAL:
Palliative Medicine


HOME HELP FEEDBACK BROWSE ARTICLES BROWSE BY AUTHOR